Marrow transplantation have taught us that hematopoiesis is supported by multipotent hematopoietic stem cells that can maintain or reconstitute the various blood cell lineages throughout life. Despite advances in our understanding of lineage commitment, this has not translated into improved approaches for treating lineage specific cytopenias. The goal of our application is to further our understanding of lineage commitment and use this knowledge to develop molecular interventions that will drive hematopoiesis toward desired lineages both in vitro and in vivo. The application is specifically focused on megakaryopoiesis, because of the clinical importance of post-transplant thrombocytopenia, and to maximize synergy with our UOl partners at Childrens Hospital of Philadelphia. Towards this goal we propose a multifaceted program with 4 Aims, with each aim involving two or more investigators from the FHCRC/UW Consortium. In Aim 1, Drs Fero, Paddington and Torok-Storb will generate molecular profiles of functionally defined progenitors in the stem cell to platelet pathway from mouse, dog, and man. These three experimental models have complementary strengths that allow for a comprehensive approach including a robust preclinical in vivo model that can predict clinical outcomes. In Aim 2 these same investigators will gene modify the defined progenitor cells to express lineage-stage-specific reporters for use in a high content, high through put siRNA screening assays to identify changes in the microenvironment that will control progenitor fate. A third Aim will test the proliferation and differentation potential of distinct progenitor subsets in response to conditionally activated signaling molecules. For this purpose Drs Blau and Emery will use signaling molecule derivatives that can be activated in response to small molecule drugs called chemical inducers of dimerization (CIDs). In Aim 4, Drs Kiem and Blau will exploit the canine model of thrombocytopenia to test gene products identified in Aim 2, and cell products expanded in Aim 3 to improve the platelet count in vivo. These in vivo studies in the dog model will establish the safety and efficacy of these therapies in a highly relevant preclinical model